Merge pull request #148 from Ensembl/feat/expand_metadata

Expand `genomes` results to include assembly and linked objects

.github/workflows/tests.yml

@@ -1,5 +1,5 @@
 # Workflow name
-name: PyTest, Black and Pylint
+name: PyTest, Black, Pylint and Mypy
 
 # Controls when the workflow will run
 on:
@@ -17,7 +17,7 @@ jobs:
   # This workflow contains a single job called "tests"
   tests:
     # The type of runner that the job will run on and timeout in minutes
-    name: Run Python Tests, Black formatter and Pylint
+    name: Run Python Tests, Black formatter, Pylint and Mypy
     runs-on: ubuntu-latest
     timeout-minutes: 10
 
@@ -32,7 +32,7 @@ jobs:
       - name: Check out repository code
         uses: actions/checkout@v3
 
-    # Set up Python version from the matrix
+      # Set up Python version from the matrix
       - name: Set up Python ${{ matrix.python-version }}
         uses: actions/setup-python@v4
         with:
@@ -59,3 +59,8 @@ jobs:
       - name: Run Pylint
         run: |
           pylint $(git ls-files '*.py') --fail-under=9.5
+
+      # Run Mypy
+      - name: Run Mypy type checker
+        run: |
+          mypy graphql_service

.gitignore

@@ -64,6 +64,7 @@ venv/
 ENV/
 env.bak/
 venv.bak/
+node_modules/
 appenv/
 
 # editor cruft

README.md

@@ -21,11 +21,16 @@ To install dependencies, run:
 Rename example_connections.conf to connections.conf and update the config values accordingly.
 
 This command will start the server:
+```
+uvicorn --workers 1 --host=0.0.0.0 graphql_service.server:APP
+```
 
-```uvicorn --workers 1 --host=0.0.0.0 graphql_service.server:APP```
-
+To run a Uvicorn server with automatic reload for development purposes, you can use the --reload flag. This flag will make Uvicorn watch your code for changes and automatically restart the server when it detects any changes.
+```
+uvicorn --workers 1 --host 0.0.0.0 --reload graphql_service.server:APP
+```
 
-If you're developing in PyCharm, you will probably find it useful to create a run 
+Also, if you're developing in PyCharm, you will probably find it useful to create a run 
 configuration so that you can use the debugger.  Create a run configuration that 
 looks like this:
 

common/db.py

@@ -17,6 +17,8 @@
 import grpc
 from ensembl.production.metadata.grpc import ensembl_metadata_pb2_grpc
 
+from common.utils import process_release_version
+
 logger = logging.getLogger(__name__)
 
 
@@ -33,26 +35,30 @@ def __init__(self, config):
         self.config = config
         self.mongo_client = MongoDbClient.connect_mongo(self.config)
 
-    def get_database_conn(self, grpc_model, uuid):
+    def get_database_conn(self, grpc_model, uuid, force_grpc=False):
         grpc_response = None
         chosen_db = self.config.get("mongo_default_db")
         # Try to connect to gRPC
         try:
             grpc_response = grpc_model.get_release_by_genome_uuid(uuid)
         except Exception as grpc_exp:
-            # chosen_db value will fall back to the default value, which is 'mongo_default_db' that is in the config
             # TODO: check why "except graphql.error.graphql_error.GraphQLError as grpc_exp:" didn't catch the error
             logger.debug(
                 "[get_database_conn] Couldn't connect to gRPC Host: %s", grpc_exp
             )
 
-        if grpc_response:
-            logger.debug("[get_database_conn] grpc_response: %s", grpc_response)
-            # replacing '.' with '_' to avoid
-            # "pymongo.errors.InvalidName: database names cannot contain the character '.'" error ¯\_(ツ)_/¯
-            release_version = str(grpc_response.release_version).replace(".", "_")
-            logger.debug("[get_database_conn] release_version: %s", release_version)
-            chosen_db = "release_" + release_version
+        if force_grpc:
+            chosen_db = process_release_version(grpc_response)
+        else:
+            if grpc_response and grpc_response.release_version:
+                chosen_db = process_release_version(grpc_response)
+            else:
+                # chosen_db value will fall back to the default value, which is 'mongo_default_db' that is in the config
+                # if force_grpc is not True
+                logger.warning(
+                    "[get_database_conn] Falling back to the default Mongo DB: '%s'",
+                    chosen_db,
+                )
 
         logger.debug("[get_database_conn] Connected to '%s' MongoDB", chosen_db)
         data_database_connection = self.mongo_client[chosen_db]
@@ -68,8 +74,8 @@ def connect_mongo(config):
         password = config.get("mongo_password")
 
         client = pymongo.MongoClient(
-            host,
-            port,
+            host=host,
+            port=port,
             username=user,
             password=password,
             read_preference=pymongo.ReadPreference.SECONDARY_PREFERRED,

common/schemas/dataset.graphql

@@ -0,0 +1,11 @@
+type Dataset {
+  dataset_id: String!
+  name: String!
+  release: Float!
+  type: String!
+  source: String!
+  dataset_type: String!
+  version: String
+  release_date: String!
+  release_type: String!
+}

common/schemas/genome.graphql

@@ -31,9 +31,12 @@ type Genome {
   assembly_accession: String!
   scientific_name: String!
   release_number: Float!
+  release_date: String!
   taxon_id: Int!
   tol_id: String
   parlance_name: String
   genome_tag: String
   is_reference: Boolean!
+  assembly: Assembly
+  dataset: [Dataset]
 }

common/schemas/query.graphql

@@ -18,15 +18,15 @@ type Query {
                  by_slice: SliceInput): Locus
   region(by_name: RegionNameInput!): Region
 
-  genomes(by_keyword: GenomeByKeywordInput,
-          by_assembly_accession_id: AssemblyAccessionIDInput): [Genome]
+
+  genomes(by_keyword: GenomeBySpecificKeywordInput): [Genome]
 
   genome(by_genome_uuid: GenomeUUIDInput!): Genome
 
 }
 
 input SymbolInput {
-  symbol: String!
+  symbol: String
   genome_id: String!
 }
 
@@ -58,9 +58,16 @@ input GenomeUUIDInput {
   release_version: Float
 }
 
-input GenomeByKeywordInput {
-  keyword: String!
-  release_version: Float
+input GenomeBySpecificKeywordInput{
+    tolid: String
+    assembly_accession_id: String
+    assembly_name: String
+    ensembl_name: String
+    common_name: String
+    scientific_name: String
+    scientific_parlance_name: String
+    species_taxonomy_id: String
+    release_version: Float
 }
 
 input AssemblyAccessionIDInput {

common/utils.py

@@ -12,6 +12,13 @@
    limitations under the License.
 """
 
+import logging
+from typing import List
+
+from graphql import GraphQLResolveInfo, FieldNode
+
+logger = logging.getLogger(__name__)
+
 
 def check_config_validity(config):
     mandatory_fields = [
@@ -30,6 +37,28 @@ def check_config_validity(config):
             )
 
 
+def process_release_version(grpc_response):
+    """
+    Processes the release version from the gRPC response and formats it for use as a database name.
+
+    This function extracts the release version from the provided gRPC response, replaces any dots ('.')
+    with underscores ('_') to avoid pymongo.errors.InvalidName errors, and returns a formatted database
+    name string.
+
+    Args:
+        grpc_response: The gRPC response object containing the release version.
+
+    Returns:
+        str: A formatted string suitable for use as a database name, prefixed with 'release_'.
+    """
+    logger.debug("[get_database_conn] grpc_response: %s", grpc_response)
+    # replacing '.' with '_' to avoid
+    # "pymongo.errors.InvalidName: database names cannot contain the character '.'" error ¯\_(ツ)_/¯
+    release_version = str(grpc_response.release_version).replace(".", "_")
+    logger.debug("[get_database_conn] release_version: %s", release_version)
+    return "release_" + release_version
+
+
 def get_ensembl_metadata_api_version():
     """
     Get the Metadata API tag from requirement.txt file
@@ -43,3 +72,29 @@ def get_ensembl_metadata_api_version():
                 version = line.strip().split("@")[-1]
                 break
     return version
+
+
+def check_requested_fields(info: GraphQLResolveInfo, fields: List[str]) -> List[bool]:
+    """
+    Check if specific fields are requested in the GraphQL query.
+
+    Args:
+        info (GraphQLResolveInfo): The GraphQL resolve information containing query details.
+        fields (List[str]): A list of field names to check for in the query.
+
+    Returns:
+        List[bool]: A list of booleans indicating whether each field is present in the query.
+
+    Usage example:
+        fields_to_check = ["assembly", "dataset"]
+        is_assembly_present, is_dataset_present = check_requested_fields(info, fields_to_check)
+    """
+    requested_fields = []
+    if info.field_nodes:
+        selection_set = info.field_nodes[0].selection_set
+        if selection_set and selection_set.selections:
+            for field in selection_set.selections:
+                if isinstance(field, FieldNode) and field.name and field.name.value:
+                    requested_fields.append(field.name.value)
+
+    return [field in requested_fields for field in fields]

graphql_service/resolver/exceptions.py

@@ -13,6 +13,7 @@
 """
 from typing import Optional, Dict
 
+import grpc
 from graphql import GraphQLError
 
 
@@ -22,11 +23,22 @@ class DatabaseNotFoundError(GraphQLError):
     """
 
     def __init__(self, db_name: str):
-        self.extensions = {"code": f"DATABASE_NOT_FOUND"}
+        self.extensions = {"code": "DATABASE_NOT_FOUND"}
         message = f"Failed to find database: {db_name}"
         super().__init__(message, extensions=self.extensions)
 
 
+class CollectionNotFoundError(GraphQLError):
+    """
+    Custom error to be raised if collection is not found
+    """
+
+    def __init__(self, collection_name: str):
+        self.extensions = {"code": "COLLECTION_NOT_FOUND"}
+        message = f"Failed to find collection: {collection_name}"
+        super().__init__(message, extensions=self.extensions)
+
+
 class FieldNotFoundError(GraphQLError):
     """
     Custom error to be raised if a field cannot be found by id
@@ -141,6 +153,15 @@ def __init__(self, organism_id):
         super().__init__("organism", {"organism_id": organism_id})
 
 
+class AssembliesFromGenomeNotFound(FieldNotFoundError):
+    """
+    Custom error to be raised if we can't find the assemblies for a genome
+    """
+
+    def __init__(self, assembly_id):
+        super().__init__("assemblies", {"assembly_id": assembly_id})
+
+
 class AssembliesFromOrganismNotFound(FieldNotFoundError):
     """
     Custom error to be raised if we can't find the assemblies for an organism
@@ -211,3 +232,17 @@ def __init__(self, message: str):
             message: The error message describing the missing argument.
         """
         super().__init__(message)
+
+
+class FailedToConnectToGrpc(grpc.RpcError):
+    """
+    Exception raised when there is gRPC connection issue.
+    """
+
+    def __init__(self, message: str):
+        """Initializes a FailedToConnectToGrpc instance.
+
+        Args:
+            message: The error message describing the issue.
+        """
+        super().__init__(message)